Respiratory system disease prediction method and system

Abstract

The invention relates to the technical field of respiratory medicine diagnosis, in particular to a respiratory system disease prediction method and system. By combining multiple groups of physicochemical data and multiple groups of image group data, comprehensive disease prediction of patients at different age stages can be realized, and compared with single-factor judgment or single-type factor judgment in the prior art, the prediction result is more accurate; clinical doctors can accurately and rapidly carry out early diagnosis on patients.

Description

Technical Field

[0001] This application relates to the technical field of respiratory medicine diagnosis, and specifically provides a respiratory disease prediction method and system. Background Art

[0002] Mycoplasma pneumoniae (MP) is one of the important pathogens of community-acquired pneumonia (CAP). The main affected population is children, but there are also occasional clinical cases in adults, mainly for adults with relatively poor resistance. It accounts for 10%-40% of CAP in children. Due to reasons such as MP's resistance to macrolide antibacterial drugs, the body's overly strong immune inflammatory response to MP, mixed infections, and hypercoagulable state, some Mycoplasma pneumoniae pneumonia (MPP) can progress to refractory Mycoplasma pneumoniae pneumonia (RMPP), and even be accompanied by intrapulmonary complications such as necrotizing pneumonia, atelectasis, plastic bronchitis, and obliterative bronchiolitis and / or damage to extrapulmonary organs, seriously endangering the physical and mental health of patients and increasing the disease burden. Early identification of RMPP remains the focus and difficulty in clinical work, but the existing clinical protocols are not sufficient for clinicians to screen out RMPP in a timely manner. Therefore, it is particularly important to explore methods with high specificity and sensitivity for the early diagnosis of RMPP. Summary of the Invention

[0003] In order to solve the technical problems existing in the prior art, the embodiment of this application provides a respiratory disease prediction method based on machine learning, which realizes the automatic prediction of respiratory diseases by combining physical and chemical index data with radiomics based on a nomogram.

[0004] In order to achieve the above purpose, the technical solution adopted in the embodiment of this application is as follows:

[0005] In a first aspect, a method for predicting respiratory diseases is provided. The method includes: obtaining a plurality of radiomic feature data corresponding to the target region image of the patient to be diagnosed, and obtaining a plurality of physical and chemical index data and a plurality of clinical data of the patient to be diagnosed; the target region includes the non-uniform normalization feature of the run length of the 2D image gray level run length matrix, the complexity feature of the adjacent gray level difference matrix of the slope image, the first-order minimum eigenvalue based on the LL filter type, the maximum two-dimensional diameter, the cluster significance of the gray level co-occurrence matrix feature, and the non-uniformity of the feature size region of the gray level size region matrix; respectively obtaining a plurality of sub-scores corresponding to the plurality of physical and chemical index data, the plurality of clinical data, and the plurality of radiomic feature data based on a first joint prediction model; the joint prediction model is a nomogram model; obtaining a total score regarding the patient to be diagnosed based on the plurality of sub-scores, and obtaining the disease risk of the patient to be diagnosed for the respiratory disease based on the score-disease risk mapping relationship.

[0006] Further, the plurality of physical and chemical index data includes the neutrophil-to-lymphocyte ratio, the serum amyloid A content value, and the lactate dehydrogenase content value.

[0007] Further, the plurality of clinical data includes the age of the patient to be diagnosed and the duration of fever of the patient to be diagnosed.

[0008] Further, the obtaining of the plurality of sub-scores corresponding to the plurality of physical and chemical index data based on the joint prediction model includes: respectively obtaining a plurality of sub-scores corresponding to the neutrophil-to-lymphocyte ratio, the serum amyloid A, and the lactate dehydrogenase content value based on the data-score mapping relationship in the nomogram model.

[0009] Further, the obtaining of the plurality of sub-scores corresponding to the plurality of clinical data based on the joint prediction model includes: respectively obtaining a plurality of sub-scores corresponding to the age data of the patient to be diagnosed and the duration of fever of the patient to be diagnosed based on the data-score mapping relationship in the nomogram model.

[0010] Further, the obtaining of the plurality of sub-scores corresponding to the plurality of radiomic feature data based on the joint prediction model includes: obtaining the radiomic scores corresponding to the plurality of radiomic feature data, and obtaining the sub-scores corresponding to the radiomic scores based on the score mapping relationship in the nomogram model.

[0011] Further, obtaining the radiomics scores corresponding to the multiple radiomics feature data includes: obtaining the scores corresponding to each feature data based on the weights corresponding to the run length non-uniform normalization feature of the 2D image gray level run length matrix, the complexity feature of the adjacent gray level difference matrix of the slope image, the first-order minimum eigenvalue based on the LL filter type, the maximum two-dimensional diameter, the cluster prominence of the gray level co-occurrence matrix feature cluster, and the size zone non-uniformity of the gray level size zone matrix feature. The calculation process is represented by the following formula: Radscores = 2.12160963720388 * 2D_glrlm_RunLengthNonUniformityNormalized + 1.75267933364089 * gradient_ngtdm_Complexity + 1.44073147131954 * wavelet-LL_firstorder_Minimum + 1.39654536708334 * shape2D_MaximumDiameter + 1.02220524269865 * glcm_ClusterProminence + 0.836846843205668 * glszm_SizeZoneNonUniformity; where 2D_glrlm_RunLengthNonUniformityNormalized represents the run length non-uniform normalization feature of the 2D image gray level run length matrix, gradient_ngtdm_Complexity represents the complexity feature of the adjacent gray level difference matrix of the slope image, wavelet-LL_firstorder_Minimum represents the first-order minimum eigenvalue based on the LL filter type, shape2D_MaximumDiameter represents the maximum two-dimensional diameter, glcm_ClusterProminence represents the cluster prominence of the gray level co-occurrence matrix feature cluster, and glszm_SizeZoneNonUniformity represents the size zone non-uniformity of the gray level size zone matrix feature.

[0012] Further, the method includes: obtaining the age distribution of the patient to be diagnosed; when the age distribution exceeds a preset age range, obtaining a plurality of radiomics feature data corresponding to the target region image of the patient to be diagnosed, and obtaining a plurality of physical and chemical index data of the patient to be diagnosed; the plurality of physical and chemical index data includes neutrophil-to-lymphocyte ratio, serum amyloid A content value, lactate dehydrogenase content value, lymphocyte value, and D-dimer value; respectively obtaining a plurality of sub-scores corresponding to the plurality of physical and chemical index data, the plurality of clinical data, and the plurality of radiomics feature data based on a second joint prediction model; obtaining an overall score for the patient to be diagnosed based on the plurality of sub-scores, and obtaining the disease risk of the patient to be diagnosed for the respiratory system disease based on the score-disease risk mapping relationship.

[0013] In a second aspect, a respiratory system disease prediction system is provided. The system includes: a data acquisition unit configured to obtain a plurality of radiomics feature data corresponding to the target region image of the patient to be diagnosed, and obtain a plurality of physical and chemical index data and a plurality of clinical data of the patient to be diagnosed; a score calculation unit configured to respectively obtain a plurality of sub-scores corresponding to the plurality of physical and chemical index data, the plurality of clinical data, and the plurality of radiomics feature data based on a first joint prediction model; and a risk calculation unit configured to obtain an overall score for the patient to be diagnosed based on the plurality of sub-scores, and obtain the disease risk of the patient to be diagnosed for the respiratory system disease based on the score-disease risk mapping relationship.

[0014] Further, the system further includes: an update unit configured to obtain the age distribution of the patient to be diagnosed; when the age distribution exceeds a preset age range, update the data acquisition unit and the score calculation unit, and obtain the disease risk of the respiratory system disease according to the data obtained by the updated data acquisition unit and the updated score calculation unit based on a second joint prediction model.

[0015] In the technical solution provided by the embodiments of the present application, by jointly using multiple sets of physical and chemical data and multiple sets of radiomics data, comprehensive disease prediction for patients of different age stages can be achieved. Compared with the judgment of a single factor or a single type of factor in the prior art, the prediction result is more accurate, enabling clinicians to diagnose patients earlier, more accurately, and quickly. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the technical solutions in the embodiments of the present application, the drawings required for the description of the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present application, and those of ordinary skill in the art can also obtain other drawings based on these drawings without creative efforts.

[0017] The methods, systems, and / or programs in the accompanying drawings will be further described according to exemplary embodiments. These exemplary embodiments will be described in detail with reference to the drawings. These exemplary embodiments are non-limiting exemplary embodiments, where example numbers represent similar mechanisms in various views of the drawings.

[0018] Figure 1 It is a schematic flowchart of a method for predicting respiratory diseases provided by an embodiment of the present application.

[0019] Figure 2 It is a schematic diagram of a first combined prediction model provided by an embodiment of the present application.

[0020] Figure 3 It is a schematic structural diagram of a system for predicting respiratory diseases provided by an embodiment of the present application.

[0021] Figure 4 It is a schematic flowchart of another method for predicting respiratory diseases provided by an embodiment of the present application.

[0022] Figure 5 It is a schematic structural diagram of another system for predicting respiratory diseases provided by an embodiment of the present application.

[0023] Figure 6 It is a schematic structural diagram of a terminal device provided by an embodiment of the present application. Detailed implementation manners

[0024] To better understand the above technical solutions, the technical solutions of the present application will be described in detail below through the accompanying drawings and specific embodiments. It should be understood that the specific features in the embodiments of the present application and the embodiments are detailed descriptions of the technical solutions of the present application, rather than limitations on the technical solutions of the present application. Without conflict, the technical features in the embodiments of the present application and the embodiments can be combined with each other.

[0025] In the following detailed description, many specific details are set forth by way of example in order to provide a thorough understanding of the relevant teachings. However, it will be apparent to those skilled in the art that the present application may be practiced without these details. In other instances, well-known methods, procedures, systems, components, and / or circuits have been described at a relatively high level without detail in order to avoid unnecessarily obscuring aspects of the present application.

[0026] Flowcharts are used in the present application to illustrate the execution processes performed by the systems according to the embodiments of the present application. It should be clearly understood that the execution processes of the flowcharts may not be executed in sequence. On the contrary, these execution processes may be executed in reverse order or simultaneously. Additionally, at least one other execution process may be added to the flowchart. One or more execution processes may be deleted from the flowchart.

[0027] Before further elaborating on the embodiments of the present invention, the nouns and terms involved in the embodiments of the present invention are described. The nouns and terms involved in the embodiments of the present invention are applicable to the following explanations.

[0028] (1) Responsive to, which is used to represent the conditions or states on which the executed operations depend. When the dependent conditions or states are met, one or more executed operations can be real-time or can have a set delay; without special instructions, there is no restriction on the execution order of the multiple executed operations.

[0029] (2) Based on, which is used to represent the conditions or states on which the executed operations depend. When the dependent conditions or states are met, one or more executed operations can be real-time or can have a set delay; without special instructions, there is no restriction on the execution order of the multiple executed operations.

[0030] Refer to Figure 1 , in the embodiments of the present application, in order to achieve the technical effects in the background art and realize the prediction of respiratory diseases, especially refractory mycoplasma pneumonia, through the combination of multiple data, a respiratory disease prediction method is provided, including the following steps:

[0031] Step S11. Obtain a plurality of radiomic feature data corresponding to the target area image of the patient to be diagnosed, and obtain a plurality of physical and chemical index data and a plurality of clinical data of the patient to be diagnosed.

[0032] In this embodiment, for the target area image being a CT image of the lung area of the patient to be diagnosed, during the acquisition process, the patient to be diagnosed undergoes a CT examination in the supine position, first with the head entering first, and then scanning in the breath-holding manner. The scanning parameters are used: 120 kV; 100 - 250 mAs; slice thickness 5 mm; pitch 1 - 1.5; matrix, 512×512, without contrast agent enhancement. All images are reconstructed using high-resolution algorithms and conventional algorithms, and the reconstructed slice thickness is 1 - 2 mm.

[0033] The plurality of radiomic feature data collected includes the non-uniform normalization feature of the run length of the 2D image gray level run length matrix, the complexity feature of the adjacent gray level difference matrix of the slope image, the first-order minimum eigenvalue based on the LL filter type, the maximum two-dimensional diameter, the cluster significance of the gray level co-occurrence matrix feature, and the non-uniformity of the feature size region of the gray level size region matrix.

[0034] For the acquisition of the above-mentioned radiomic feature data in this embodiment, an existing set of diseased images is selected, and 1032 radiomic features are automatically extracted from each ROI using PyRadiomics software. The feature values of each dimension are stretched between [0, 1] using min-max normalization for standardization. Then, the maximum relevance minimum redundancy algorithm is used to select features with high correlation. Next, LASSO is used to further screen the features, and the most suitable λ value is selected to retain the highly correlated feature subset. Five-fold cross-validation is performed to avoid overfitting of the LASSO model. The final features are input into a support vector machine machine learning model for the construction of a prediction model, and the radiomic score of each patient is calculated, and the scores are statistically screened. For statistical methods, SPSS 26.0 statistical software and R language (version 4.3.2) are used. Data conforming to a normal distribution are expressed as mean ± standard deviation, and independent t-tests are used for comparative analysis; skewed distribution data are expressed as median (percentile). Categorical variables are expressed as absolute numbers (n) and proportions (%), and chi-square tests or Fisher's exact tests are used for comparative analysis. Finally, a total of 6 optimal radiomic features are screened through LASSO regression, and an ROC curve of the prediction efficacy of the radiomic label is established. Correlation combination analysis is performed on the 6 extracted radiomic features, and the radiomic feature scores of patients are calculated according to the regression coefficients.

[0035] In this embodiment, the multiple physical and chemical index data include the neutrophil-to-lymphocyte ratio, the serum amyloid A content value, and the lactate dehydrogenase content value. Among them, under the stimulation of infection or inflammation, inflammatory factors produced by the body, such as IL-1β, IL-6, etc., will stimulate hepatocytes to synthesize a large amount of SAA, promote the production of the above cytokines and chemokines such as CCL2 and CXCL8, thereby amplifying the inflammatory response. Through research, SAA in children with severe MPP is significantly higher than that in children with non-severe MPP. The applicant found that in the clinical data, the SAA level in the RMPP group of children is significantly higher than that in the non-RMPP group of children. The applicant found through multivariate logistic regression analysis that SAA is an independent risk factor for the occurrence of RMPP. Although in the prior art, the relationship between the change in SAA level and RMPP has been clear, different from the prior art, in this embodiment, the SAA index is combined with the NLR index and the LDH index, and the combined effect can better reflect the level change of RMPP. The NLR index represents the ratio of peripheral blood neutrophil to lymphocyte count, which is a new biomarker that can reflect both innate immune response and adaptive immunity, and has been used to predict the poor prognosis of various infectious diseases. After MP infection, neutrophils increase in peripheral blood, BALF and lung tissue, playing a role in local killing and phagocytosis of pathogens. However, excessive aggregation and activation of neutrophils will lead to the inflammatory cascade effect and immune imbalance. At the same time, excessive inflammation can induce lymphocyte apoptosis, resulting in a decrease in the number of lymphocytes. The applicant found that for children in the RMPP group, the peripheral blood N% is higher, the L% is lower, and the NLR level is higher. Multivariate regression analysis confirmed that NLR can be used as an independent predictor for the occurrence of RMPP, and high NLR is the common result of an increase in neutrophil count and a decrease in lymphocyte count, which can better reflect the systemic inflammatory state of the body. LDH is an inflammatory marker and plays an important role in the glycolytic pathway. LDH is widely present in tissue cells, and its activity in lung tissue is second only to that in tissues such as the kidney, myocardium, and liver, much higher than that in blood. Therefore, even a small amount of lung tissue damage can cause changes in the serum LDH concentration. During pneumonia, due to lung inflammation and tissue hypoxia, the cell membrane permeability increases, and the enzymes in lung tissue cells are released into the blood, increasing the LDH level in the blood, which can reflect the severity of the disease.

[0036] By combining the above three indexes in this embodiment, the prediction of RMPP can be better achieved, and it has higher sensitivity and specificity compared with the commonly used index substances in the prior art.

[0037] For multiple clinical data including the age of the patient to be diagnosed and the duration of fever of the patient to be diagnosed, these two data can reflect the degree of the patient's clinical response and are direct clinical evidence for indicating the current probability of illness.

[0038] Step S12. Based on the first combined prediction model, obtain multiple sub-scores corresponding to the multiple physical and chemical index data, the multiple clinical data, and the multiple radiomic feature data respectively.

[0039] In this embodiment, the first combined prediction model is a nomogram model. For this nomogram model, reference can be made to Figure 2 As shown, it includes the score intervals corresponding to the distribution of each index data, and the corresponding sub-score can be determined according to the current numerical distribution of each index data. This method is commonly used in existing disease prediction and evaluation schemes and will not be elaborated in this embodiment.

[0040] For the acquisition of each sub-score, based on the household-number - score mapping relationship in the nomogram model, obtain multiple sub-scores corresponding to the neutrophil-to-lymphocyte ratio, serum amyloid A, and lactate dehydrogenase content values respectively, as well as multiple sub-scores corresponding to the age data of the patient to be diagnosed and the fever duration of the patient to be diagnosed.

[0041] For the sub-score corresponding to the radiomic data, first determine the weights corresponding to the multiple radiomic feature data in step S11, perform weighted summation to obtain the radiomic score, and then obtain the sub-score corresponding to the radiomic score based on the score mapping relationship in the nomogram model.

[0042] Among them, the calculation method of the radiomics score is expressed based on the following formula: Radscores = 2.12160963720388 * 2D_glrlm_RunLengthNonUniformityNormalized + 1.75267933364089 * gradient_ngtdm_Complexity + 1.44073147131954 * wavelet-LL_firstorder_Minimum + 1.39654536708334 * shape2D_MaximumDiameter + 1.02220524269865 * glcm_ClusterProminence + 0.836846843205668 * glszm_SizeZoneNonUniformity; where 2D_glrlm_RunLengthNonUniformityNormalized represents the feature of the run length non-uniform normalization based on the 2D image gray level run length matrix, gradient_ngtdm_Complexity represents the complexity feature of the adjacent gray level difference matrix of the slope image, wavelet-LL_firstorder_Minimum represents the first-order minimum eigenvalue based on the LL filter type, shape2D_MaximumDiameter represents the maximum two-dimensional diameter, glcm_ClusterProminence represents the feature cluster prominence of the gray level co-occurrence matrix, and glszm_SizeZoneNonUniformity represents the feature size zone non-uniformity of the gray level size zone matrix.

[0043] Step S13. Obtain the total score for the patient to be diagnosed based on multiple sub-scores, and obtain the disease risk of the patient to be diagnosed for the respiratory system disease based on the score-disease risk mapping relationship.

[0044] See Figure 2 It can be seen that in the nomogram model, there is also a score-disease risk mapping relationship, and through this relationship, the total score calculated in step S12 can be converted into a disease risk.

[0045] For the embodiment of the present application, a respiratory system disease prediction method is provided. By combining clinical data, physical and chemical index data, and radiomics feature data and based on the nomogram model, the refractory mycoplasma pneumonia can be accurately predicted at an early stage.

[0046] See Figure 3, based on steps S11 - S13, a virtual system, namely the respiratory disease prediction system 30, is also provided for performing the processing procedures of steps S11 - S13. Regarding this system, it includes the following units:

[0047] The data acquisition unit 31 is used to acquire multiple radiomics feature data corresponding to the target region image of the patient to be diagnosed, as well as acquire multiple physical and chemical index data and multiple clinical data of the patient to be diagnosed;

[0048] The score calculation unit 32 is used to respectively obtain multiple sub - scores corresponding to the multiple physical and chemical index data, the multiple clinical data, and the multiple radiomics feature data based on the first combined prediction model;

[0049] The risk calculation unit 33 is used to obtain the total score of the patient to be diagnosed based on multiple sub - scores, and obtain the disease risk of the patient to be diagnosed for the respiratory disease based on the score - disease risk mapping relationship.

[0050] A respiratory disease prediction method provided for steps S11 - S13 can achieve early prediction of most refractory mycoplasma pneumonia in clinical scenarios. However, it should be noted that since mycoplasma pneumoniae is commonly found in children clinically, the prediction method in steps S11 - S13 is applicable to most clinical scenarios, that is, the children's disease scenario. Although the number of adult patients is small, it also exists clinically. And there are differences in clinical data and physical and chemical data between children and adult patients. Therefore, in order to further improve the clinical application scope of this prediction method. In this embodiment, based on the basic principles of steps S11 - S13, another respiratory disease prediction method is also provided for predicting the likelihood of adult RMPP.

[0051] Specifically, refer to Figure 4 , regarding this respiratory disease prediction method, it includes the following steps:

[0052] Step S41. Obtain the age distribution of the patient to be diagnosed. When the age distribution exceeds the pre - set age range, obtain multiple radiomics feature data corresponding to the target region image of the patient to be diagnosed, and obtain multiple physical and chemical index data of the patient to be diagnosed.

[0053] In this embodiment, the screening for children and adults is judged by the age of the patient to be diagnosed. When the age of the patient to be diagnosed exceeds the pre - set value of the children's age, it indicates that the patient to be diagnosed is an adult. Specifically, the pre - set value in this embodiment is 16, that is, when it exceeds 16 years old, it indicates that this patient is a non - child patient.

[0054] Among them, for this embodiment, when exceeding this preset value, different from the previous embodiment, in this embodiment, clinical data is no longer collected, but only multiple physical and chemical index data and multiple radiomics feature data are collected. Moreover, there are also differences in the physical and chemical index data from the previous embodiment. In this embodiment, the physical and chemical index data includes the neutrophil-to-lymphocyte ratio, serum amyloid A content, lactate dehydrogenase content, lymphocyte value, and D-dimer value. The radiomics feature data is the same as that in the previous embodiment, and also includes the non-uniform normalized feature of the run length of the gray level run length matrix based on 2D images, the complexity feature of the adjacent gray level difference matrix of the slope image, the first-order minimum eigenvalue based on the LL filter type, the maximum two-dimensional diameter, the cluster significance of the gray level co-occurrence matrix feature, and the non-uniformity of the feature size region of the gray level size region matrix.

[0055] Step S42. Based on the second combined prediction model, respectively obtain multiple sub-scores corresponding to the multiple physical and chemical index data, the multiple clinical data, and the multiple radiomics feature data.

[0056] Step S43. Obtain the total score regarding the patient to be diagnosed based on the multiple sub-scores, and obtain the disease risk of the patient to be diagnosed for the respiratory system disease based on the score-disease risk mapping relationship.

[0057] In this embodiment, the second combined prediction model is also a nomogram model, but different from the first combined prediction model, in this embodiment, due to the change in the physical and chemical index data, there is no longer a mapping relationship corresponding to the clinical data in the second combined prediction model, and the mapping relationships corresponding to the lymphocyte value and the D-dimer value are also added.

[0058] Regarding the method for obtaining the sub-scores and the total score in this process is the same as steps S12 and S13, that is, determining the sub-score corresponding to each data through the data-score mapping relationship, and obtaining the total score by summing the sub-scores. This process will not be elaborated in this embodiment.

[0059] Refer to Figure 5 , for steps S41 - S43, which are updates and changes to steps S11 - S13, and for this updated method, in this embodiment, the respiratory system disease prediction system 30 is also updated. The updated respiratory system disease prediction system 50 includes the following units:

[0060] A data acquisition unit 51, configured to acquire multiple radiomics feature data corresponding to the target region image of the patient to be diagnosed, as well as acquire multiple physical and chemical index data and multiple clinical data of the patient to be diagnosed;

[0061] A score calculation unit 52 is configured to respectively obtain multiple sub-scores corresponding to the multiple physical and chemical index data, the multiple clinical data, and the multiple radiomic feature data based on the first joint prediction model.

[0062] A risk calculation unit 53 is configured to obtain a total score for the patient to be diagnosed based on the multiple sub-scores, and obtain the disease risk of the patient to be diagnosed for the respiratory system disease based on the score-disease risk mapping relationship.

[0063] An update unit 54 is configured to obtain the age distribution of the patient to be diagnosed. When the age distribution exceeds a preset age range, the data acquisition unit and the score calculation unit are updated, and the disease risk of the respiratory system disease is obtained based on the data acquired by the updated data acquisition unit and the updated score calculation unit according to the second joint prediction model.

[0064] Refer to Figure 6 In addition, the above method can be integrated into the provided terminal device 60. Due to relatively large differences that may occur due to different configurations or performances of the device, it may include one or more processors 601 and a memory 602. One or more application programs or data may be stored in the memory 602. Among them, the memory 602 may be short-term storage or persistent storage. The application programs stored in the memory 602 may include one or more modules (not shown in the figure), and each module may include a series of computer executable instructions in the terminal device. Further, the processor 601 may be set to communicate with the memory 402, and execute a series of computer executable instructions in the memory 602 on the terminal device. The terminal device may also include one or more power supplies 603, one or more wired or wireless network interfaces 604, one or more input / output interfaces 605, one or more keyboards 606, etc.

[0065] In a specific embodiment, the terminal device includes a memory, and one or more programs, where one or more programs are stored in the memory, and one or more programs may include one or more modules, and each module may include a series of computer executable instructions in the terminal device, and is configured to be executed by one or more processors. The one or more programs include the following computer executable instructions:

[0066] Obtain multiple radiomic feature data corresponding to the target region image of the patient to be diagnosed, and obtain multiple physical and chemical index data and multiple clinical data of the patient to be diagnosed.

[0067] Obtain multiple sub-scores corresponding to the multiple physical and chemical index data, the multiple clinical data, and the multiple radiomic feature data respectively based on the first combined prediction model;

[0068] Obtain the total score of the patient to be diagnosed based on the multiple sub-scores, and obtain the disease risk of the patient to be diagnosed for the respiratory disease based on the score-disease risk mapping relationship.

[0069] Or execute the following instructions:

[0070] Obtain the age distribution of the patient to be diagnosed. When the age distribution exceeds the preset age range, obtain multiple radiomic feature data corresponding to the target region image of the patient to be diagnosed, and obtain multiple physical and chemical index data of the patient to be diagnosed;

[0071] Obtain multiple sub-scores corresponding to the multiple physical and chemical index data, the multiple clinical data, and the multiple radiomic feature data respectively based on the second combined prediction model;

[0072] Obtain the total score of the patient to be diagnosed based on the multiple sub-scores, and obtain the disease risk of the patient to be diagnosed for the respiratory disease based on the score-disease risk mapping relationship.

[0073] The following is a specific introduction to each component of the processor:

[0074] Among them, in this embodiment, the processor is an application specific integrated circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of the present application, for example: one or more digital signal processors (DSPs), or one or more field programmable gate arrays (FPGAs).

[0075] Optionally, the processor can execute various functions by running or executing software programs stored in the memory and calling data stored in the memory, such as executing the above Figure 1 Or / and Figure 4 The method shown.

[0076] In a specific implementation, as an embodiment, the processor may include one or more microprocessors.

[0077] The memory is used to store the software program for executing the solution of the present application and is controlled by the processor to execute. The specific implementation manner can refer to the above method embodiments and will not be elaborated here.

[0078] Optionally, the memory may be a read-only memory (ROM) or other type of static storage device that can store static information and instructions, a random access memory (RAM) or other type of dynamic storage device that can store information and instructions, or may also be an electrically erasable programmable read-only memory (EEPROM), a compact disc read-only memory (CD-ROM), or other optical disc storage, optical disc storage (including compact discs, laser discs, optical discs, digital versatile discs, Blu-ray discs, etc.), magnetic disk storage media, or other magnetic storage devices, or any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory may be integrated with the processor or may exist independently and be coupled to the processing unit through the interface circuit of the processor. The embodiments of the present application do not make specific limitations on this.

[0079] It should be noted that the structure of the processor shown in this embodiment does not constitute a limitation on the device. The actual device may include more or fewer components than shown in the figure, or combine some components, or have a different component layout.

[0080] In addition, the technical effects of the processor can refer to the technical effects of the method described in the above method embodiments and will not be elaborated here.

[0081] It should be understood that the processor in the embodiments of the present application may be other general-purpose processors, digital signal processors (DSPs), application specific integrated circuits (ASICs), field programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or the processor may also be any conventional processor, etc.

[0082] It should also be understood that the memory in the embodiments of the present application may be a volatile memory or a non-volatile memory, or may include both volatile and non-volatile memories. Among them, the non-volatile memory may be a read-only memory (ROM), a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), or a flash memory. The volatile memory may be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of random access memory (RAM) are available, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchlink DRAM (SLDRAM), and direct rambus RAM (DR RAM).

[0083] The above embodiments can be implemented in whole or in part by software, hardware (such as circuits), firmware, or any combination thereof. When implemented using software, the above embodiments can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions or computer programs. When the computer instructions or computer programs are loaded or executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center by wired (such as infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that contains one or more collections of available media. The available media can be magnetic media (such as floppy disks, hard disks, magnetic tapes), optical media (such as DVDs), or semiconductor media. The semiconductor media can be a solid-state drive.

[0084] In the present application, "at least one" means one or more, and "a plurality" means two or more. "At least one of the following" or a similar expression means any combination of these items, including any combination of single items or plural items. For example, at least one of a, b, or c can represent: a, b, c, a - b, a - c, b - c, or a - b - c, where a, b, and c can be single or multiple.

[0085] It should be understood that in various embodiments of the present application, the order numbers of the above processes do not indicate the order of execution. The order of execution of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application.

[0086] Those of ordinary skill in the art can realize that the units and algorithm steps of each example described in combination with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are executed in a hardware or software manner depends on the specific application and design constraints of the technical solution. A professional technician can use different methods for each specific application to implement the described functions, but such implementation should not be considered to exceed the scope of the present application.

[0087] Those skilled in the art can clearly understand that for the convenience and brevity of description, the specific working processes of the systems, devices, and units described above can refer to the corresponding processes in the foregoing method embodiments and will not be elaborated herein.

[0088] In several embodiments provided in the present application, it should be understood that the disclosed systems, devices, and methods can be implemented in other ways. For example, the device embodiments described above are merely illustrative. For example, the division of the units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the displayed or discussed couplings or direct couplings or communication connections to each other can be through some interfaces. The indirect couplings or communication connections of the devices or units can be in electrical, mechanical, or other forms.

[0089] The units described as separate components may or may not be physically separated. The components displayed as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0090] In addition, in each embodiment of the present application, the functional units can be integrated into one processing unit, or each unit can exist physically alone, or two or more units can be integrated into one unit.

[0091] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application, in essence, or the part that contributes to the prior art or a part of this technical solution can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in each embodiment of the present application. The foregoing storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM), random access memories (RAM), magnetic disks, or optical discs that can store program codes.

[0092] As described above, it is only the specific implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should all be covered within the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims described above.

Claims

1. A method for predicting respiratory diseases, characterized in that, The method includes: Obtaining a plurality of radiomics feature data corresponding to the target region image of the patient to be diagnosed, and obtaining a plurality of physical and chemical index data and a plurality of clinical data of the patient to be diagnosed; the target region image is a CT image of the lung region of the patient to be diagnosed, and the plurality of radiomics feature data include the non-uniform normalization feature of the run length of the 2D image gray level run length matrix, the complexity feature of the adjacent gray level difference matrix of the slope image, the first-order minimum eigenvalue based on the LL filter type, the maximum two-dimensional diameter, the cluster significance of the gray level co-occurrence matrix feature, and the non-uniformity of the feature size region of the gray level size region matrix; Based on the first joint prediction model, respectively obtaining a plurality of sub-scores corresponding to the plurality of physical and chemical index data, the plurality of clinical data, and the plurality of radiomics feature data; the first joint prediction model is a nomogram model; Obtaining a total score for the patient to be diagnosed based on the plurality of sub-scores, and obtaining the disease risk of the patient to be diagnosed for the respiratory system disease based on the score-disease risk mapping relationship.

2. The respiratory disease prediction method according to claim 1, characterized in that The plurality of physical and chemical index data include the neutrophil-to-lymphocyte ratio, the serum amyloid A content value, and the lactate dehydrogenase content value.

3. The respiratory disease prediction method according to claim 1, wherein The plurality of clinical data include the age of the patient to be diagnosed and the duration of fever of the patient to be diagnosed.

4. The respiratory disease prediction method according to claim 2, wherein The obtaining of the plurality of sub-scores corresponding to the plurality of physical and chemical index data based on the first joint prediction model includes: respectively obtaining a plurality of sub-scores corresponding to the neutrophil-to-lymphocyte ratio, the serum amyloid A, and the lactate dehydrogenase content value based on the data-score mapping relationship in the nomogram model.

5. The respiratory disease prediction method according to claim 3, wherein The obtaining of the plurality of sub-scores corresponding to the plurality of clinical data based on the first joint prediction model includes: respectively obtaining a plurality of sub-scores corresponding to the age data of the patient to be diagnosed and the duration of fever of the patient to be diagnosed based on the data-score mapping relationship in the nomogram model.

6. The respiratory disease prediction method according to claim 1, wherein The obtaining of the plurality of sub-scores corresponding to the plurality of radiomics feature data based on the first joint prediction model includes: obtaining the radiomics score corresponding to the plurality of radiomics feature data, and obtaining the sub-score corresponding to the radiomics score based on the score mapping relationship in the nomogram model.

7. The respiratory disease prediction method according to claim 6, wherein, Obtaining the radiomics scores corresponding to the multiple radiomics feature data includes: obtaining the score corresponding to each feature data based on the weights corresponding to the run length non-uniform normalization feature of the 2D image gray level run length matrix, the complexity feature of the adjacent gray level difference matrix of the gradient image, the first-order minimum eigenvalue based on the LL filtering type, the maximum two-dimensional diameter, the cluster prominence of the gray level co-occurrence matrix feature, and the size zone non-uniformity of the gray level size zone matrix feature. The calculation process is represented by the following formula: Radscores = 2.12160963720388 * 2D_glrlm_RunLengthNonUniformityNormalized + 1.75267933364089 * gradient_ngtdm_Complexity + 1.44073147131954 * wavelet-LL_firstorder_Minimum + 1.39654536708334 * shape2D_MaximumDiameter + 1.02220524269865 * glcm_ClusterProminence + 0.836846843205668 * glszm_SizeZoneNonUniformity; where 2D_glrlm_RunLengthNonUniformityNormalized represents the run length non-uniform normalization feature of the 2D image gray level run length matrix, gradient_ngtdm_Complexity represents the complexity feature of the adjacent gray level difference matrix of the gradient image, wavelet-LL_firstorder_Minimum represents the first-order minimum eigenvalue based on the LL filtering type, shape2D_MaximumDiameter represents the maximum two-dimensional diameter, glcm_ClusterProminence represents the cluster prominence of the gray level co-occurrence matrix feature, and glszm_SizeZoneNonUniformity represents the size zone non-uniformity of the gray level size zone matrix feature.

8. The respiratory disease prediction method according to claim 1, wherein The method includes: obtaining the age distribution of the patient to be diagnosed, when the age distribution exceeds a preset age range, obtaining multiple radiomics feature data corresponding to the target region image of the patient to be diagnosed, and obtaining multiple physical and chemical index data of the patient to be diagnosed; the multiple physical and chemical index data include the neutrophil-to-lymphocyte ratio, the serum amyloid A content value, the lactate dehydrogenase content value, the lymphocyte value, and the D-dimer value; Based on the second combined prediction model, respectively obtaining multiple sub-scores corresponding to the multiple physical and chemical index data, the multiple clinical data, and the multiple radiomics feature data; Obtaining the total score of the patient to be diagnosed based on the multiple sub-scores, and obtaining the disease risk of the patient to be diagnosed for the respiratory system disease based on the score-disease risk mapping relationship.

9. A respiratory disease prediction system, characterized in that, The system includes: a data acquisition unit, configured to acquire a plurality of radiomics feature data corresponding to the target region image of the patient to be diagnosed, and acquire a plurality of physical and chemical index data and a plurality of clinical data of the patient to be diagnosed; a score calculation unit, configured to respectively obtain a plurality of sub-scores corresponding to the plurality of physical and chemical index data, the plurality of clinical data, and the plurality of radiomics feature data based on a first joint prediction model; a risk calculation unit, configured to obtain a total score regarding the patient to be diagnosed based on the plurality of sub-scores, and obtain the disease risk of the patient to be diagnosed for the respiratory system disease based on a score-disease risk mapping relationship.

10. The respiratory disease prediction system according to claim 9, characterized in that, The system further includes: an update unit, configured to acquire the age distribution of the patient to be diagnosed, and when the age distribution exceeds a preset age range, update the data acquisition unit and the score calculation unit, and obtain the disease risk of the respiratory system disease based on the data acquired by the updated data acquisition unit and the updated score calculation unit according to a second joint prediction model.

Images